INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

utherland

-0.75

eger

-0.73

 Greenwald

-0.70

Cash

-0.69

CLUD

-0.67

cox

-0.67

INESS

-0.67

enne

-0.66

UFF

-0.66

earch

-0.66

POSITIVE LOGITS

tailed

0.80

 duty

0.70

testing

0.68

 prep

0.61

tips

0.60

arers

0.59

ecd

0.59

ewitness

0.59

 witnessing

0.58

 testing

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.